The requirement is to Plot a bar chart using a seaborn library, showing total sales (Customer_Value) by the Last_region. Here my code

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
customer = pd.read_csv('D:\PythonTraining\Customer.csv')

df = customer.groupby('Last_region')['Customer_Value'].sum()
df

sns.barplot(x = df.Last_region, y = df.Customer_Value)
plt.show

I get an error:

AttributeError: 'Series' object has no attribute 'Last_region'.

How can I correct it? I believed after groupby, the attribute cant be referenced.

CodePudding user response：

The issue is that Last_region becomes the index when you group on it. Also note that df here is most likely a Series, not a DataFrame, in which case Customer_Value would also not be a column.

• Either use x=df.index and y=df.values

  sns.barplot(x=df.index, y=df.values)
  

• Or use data=df.reset_index() (now it's guaranteed to be a DataFrame with those columns)

  sns.barplot(data=df.reset_index(), x='Last_region', y='Customer_Value')
  

Alternatively, as Asish commented, you can change df so that Last_region is not the index:

• Either set as_index=False while grouping

  df = customer.groupby('Last_region', as_index=False)['Customer_Value'].sum()
  

• Or reset_index() after grouping

  df = customer.groupby('Last_region')['Customer_Value'].sum().reset_index()